It is currently Mon Nov 11, 2019 1:34 pm

All times are UTC - 8 hours [ DST ]




Post new topic Reply to topic  [ 13 posts ] 
Author Message
Offline
 Post subject: New networks Leela Zero, a Sensible Artificial Intelligence
Post #1 Posted: Wed Oct 30, 2019 5:39 am 
Lives with ko

Posts: 158
Liked others: 31
Was liked: 10
http://sai.unich.it

Top
 Profile  
 
Offline
 Post subject: Re: New networks Leela Zero, a Sensible Artificial Intellige
Post #2 Posted: Fri Nov 01, 2019 4:22 am 
Lives with ko

Posts: 242
Liked others: 4
Was liked: 57
Very nice, thanks !

sai9x9 seems very strong, about as strong as katago1.2 ?
9x9 Sabaki games between katago and sai are often very close.
(here, both programs play around 2-5 sec/move)



sai9x9 can win against gnugo at H2 and H3, and even sometimes at H4 (komi 0.5)





Attachments:
sgh4.sgf [986 Bytes]
Downloaded 446 times
sgh3.sgf [731 Bytes]
Downloaded 438 times
sw1.sgf [1.03 KiB]
Downloaded 439 times
Top
 Profile  
 
Offline
 Post subject: Re: New networks Leela Zero, a Sensible Artificial Intellige
Post #3 Posted: Fri Nov 01, 2019 5:24 am 
Lives with ko

Posts: 158
Liked others: 31
Was liked: 10
if I understood correctly, sai9x9 uses a network 94619dea? but the network was received on 12/21/2018? almost a year ago! but strange, I haven’t seen this network before

Top
 Profile  
 
Offline
 Post subject: Re: New networks Leela Zero, a Sensible Artificial Intellige
Post #4 Posted: Mon Nov 04, 2019 10:44 am 
Lives with ko

Posts: 158
Liked others: 31
Was liked: 10
last network 848fe9be (10 sec/move, cpu version) http://sai.unich.it/networks/848fe9be1b ... 99fbf84.gz
against zen 7 1 dan (gtp4zen -t4 -s1800 -n1 -o1.3 -p0.35) - zen 7 surrendered on move 109


Attachments:
sai 848fe9be - zen 7.sgf [964 Bytes]
Downloaded 15 times
Top
 Profile  
 
Offline
 Post subject: Re: New networks Leela Zero, a Sensible Artificial Intellige
Post #5 Posted: Tue Nov 05, 2019 10:38 am 
Lives with ko

Posts: 158
Liked others: 31
Was liked: 10
last network 43268842 (10 sec/move, cpu version) http://sai.unich.it/networks/43268842e5 ... 5d364fe.gz
against ZEN 7 2 dan (gtp4zen -t4 -s2000 -n1 -o1.0 -p0.4) - SAI won +2.5 (302 moves)
20 more games SAI & ZEN - 4:16


Attachments:
sai 43268842 - zen 7.sgf [2.43 KiB]
Downloaded 10 times
Top
 Profile  
 
Offline
 Post subject: Re: New networks Leela Zero, a Sensible Artificial Intellige
Post #6 Posted: Thu Nov 07, 2019 5:54 am 
Lives with ko

Posts: 158
Liked others: 31
Was liked: 10
SAI #64 fe1cfa3c (10 sec/move, cpu version) http://sai.unich.it/networks/fe1cfa3c81 ... 95f2dfb.gz
against ZEN 7 3 dan (gtp4zen -t4 -s2200 -n2 -o2.0 -p0.45) - 2:18
zen surrendered in two games:


Attachments:
fe1cfa3c - zen 7 2.sgf [1.55 KiB]
Downloaded 11 times
fe1cfa3c - zen 7.sgf [1.46 KiB]
Downloaded 11 times
Top
 Profile  
 
Offline
 Post subject: Re: New networks Leela Zero, a Sensible Artificial Intellige
Post #7 Posted: Fri Nov 08, 2019 5:33 am 
Lives with ko

Posts: 242
Liked others: 4
Was liked: 57
Two more Sabaki 9x9 games Sai-Katago (both at 40000 playouts)
Maybe Katago 1.2 wins slightly more often than Sai





Attachments:
saikata2.sgf [752 Bytes]
Downloaded 147 times
saikata.sgf [883 Bytes]
Downloaded 144 times
Top
 Profile  
 
Offline
 Post subject: Re: New networks Leela Zero, a Sensible Artificial Intellige
Post #8 Posted: Fri Nov 08, 2019 6:14 am 
Lives with ko

Posts: 242
Liked others: 4
Was liked: 57
Two Sabaki 9x9 games Sai-gtp4zen (40000 playouts for Sai, -T10 -s40000 for Zen)
Sai wins even games and can win H2 with komi 7.5 (but not always)




Attachments:
saizen_h2_k75.sgf [612 Bytes]
Downloaded 132 times
saizen.sgf [998 Bytes]
Downloaded 139 times
Top
 Profile  
 
Offline
 Post subject: Re: New networks Leela Zero, a Sensible Artificial Intellige
Post #9 Posted: Fri Nov 08, 2019 7:10 am 
Beginner

Posts: 3
Liked others: 7
Was liked: 7
Rank: 9k
KGS: vandertic
DGS: vandertic
Hi, SAI main author here!

What options are you using with SAI?

Be aware at least of the following ones:

--lambda [0.5]
--mu [0]
--nrsymm

The latter is the simplest. Merge symmetrical positions in order to explore more deeply. Highly recommended when playing at fixed visits as it could give quite a lot of strength increase

lambda and mu are real parameters intended as 0<=mu<=lambda<=1. They tell SAI to consider a range of komi values when playing, instead of just the real komi of the game. If you are playing with an opponent of equal strength it is recommended to set both to 0 (notice that this is NOT the default for lambda).

If you are playing with a weaker opponent, and/or there is handicap, the default values should be ok, at least for 9x9


This post by frmor was liked by: And
Top
 Profile  
 
Offline
 Post subject: Re: New networks Leela Zero, a Sensible Artificial Intellige
Post #10 Posted: Fri Nov 08, 2019 10:49 am 
Lives with ko

Posts: 158
Liked others: 31
Was liked: 10
frmor, thank you so much for this project!
I correctly understood, these recommendations for the case if SAI are used to play against another program (human)?
which values are better if the opponent is stronger? and mu = 0.5 or 1 can be used, and what will it give?

Top
 Profile  
 
Offline
 Post subject: Re: New networks Leela Zero, a Sensible Artificial Intellige
Post #11 Posted: Fri Nov 08, 2019 8:32 pm 
Lives with ko

Posts: 242
Liked others: 4
Was liked: 57
frmor wrote:
--lambda [0.5]
--mu [0]
--nrsymm
and --noponder -p 40000 --komi x.x
Thank you for your detailed explanation about lambda and mu. I like SAI very much :tmbup:

Top
 Profile  
 
Offline
 Post subject: Re: New networks Leela Zero, a Sensible Artificial Intellige
Post #12 Posted: Sat Nov 09, 2019 4:19 am 
Beginner

Posts: 3
Liked others: 7
Was liked: 7
Rank: 9k
KGS: vandertic
DGS: vandertic
And wrote:
frmor, thank you so much for this project!
I correctly understood, these recommendations for the case if SAI are used to play against another program (human)?
which values are better if the opponent is stronger? and mu = 0.5 or 1 can be used, and what will it give?


It's a bit complicated, and you should read the paper to get full details. Let us say that lambda and mu control the two sides of the interval of komi values over which the winrate is averaged.

If mu=0 it means to keep the real komi as one side.

If lambda=1 it means to use, for the other side, the komi value that SAI thinks would make the current position fair (in the sense of 50% winrate).

Intermediate values are to be intended in the winrate scale, not komi scale, so that for example, if SAI thinks it is ahead, with winrate 80% and lambda=0.3, than it would compute the weighted mean 71% = 0.3 * 50% + 0.7 * 80%, look for the komi value that would give itself a 71% winrate, and then use this as the side.

So if SAI is black and real komi is 7.5, then this virtual komi value would be something more, say 9.3. Then SAI for the exploration of the current move would compute the nodes winrates as averages over the komi range [7.5 ; 9.3] and choose its move accordingly.

You can put mu=lambda>0 and in this way SAI would play as if the komi was the one computed, hence giving itself increasing score handicap when ahead. (And the opposite when behind.)



My recommendation is to use lambda=mu=0 when against a strong opponent, to try hard to win with the real komi.

You should use values larger than 0 when against a weaker opponent, both to improve the final score, when playing a fair game, and to improve the strength of play when starting with handicap.

When SAI is behind with a weaker opponent because of handicap, positive values of lambda (and maybe mu) were showed to improve its ability to recover from the initial disadvantage.


This post by frmor was liked by 2 people: And, Waylon
Top
 Profile  
 
Offline
 Post subject: Re: New networks Leela Zero, a Sensible Artificial Intellige
Post #13 Posted: Sat Nov 09, 2019 6:48 am 
Lives with ko

Posts: 158
Liked others: 31
Was liked: 10
SAI #70 c74e6c7d (10 sec/move, cpu version) http://sai.unich.it/networks/c74e6c7d84 ... 2c385ee.gz
against ZEN 7 4 dan (gtp4zen -t4 -s2400 -n2 -o1.5 -p0.5) - 1:19
zen surrendered in one game:


Attachments:
sai c74e6c7d - zen 7.sgf [1.58 KiB]
Downloaded 55 times
Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 13 posts ] 

All times are UTC - 8 hours [ DST ]


Who is online

Users browsing this forum: No registered users and 1 guest


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group